WHAT NEW IN X10.30 FRI

X 上的 AI 合集档案 · 22 个合集 · 60,637 条

gemini-4POST

质疑 Artificial Analysis 压低开源模型分数、抬高 Gemini 4 Argon

发帖者质疑 AA 榜单刻意压低开源权重模型得分,并反问 Gemini 4 Argon 在 agent 任务上是否真强于 DeepSeek V4.1 1017。

dealign.ai
@dealignai17 小时前收录原推发布于 2026.10.28
dealign.ai @dealignai原推发布于 2026.10.28
在 X 上打开原推

is it just me or does it keep feeling like AA at every chance goes out of their way to make open local models feel stupider than they are? why are they going out of their way to make the scores of open weight models lower? are you trying to tell me gemini 4 argon is better to use for agentic shit than dsv41 1017?

这条是引用着另一条写的:x.com/aibattle_/status/2115269739069026766

x.com/dealignai/status/2115280601456705705采集于 2026.10.29 10:07